On 2025-04-04 at 20:48:07 UTC-0400 (Fri, 4 Apr 2025 18:48:07 -0600)
Tom Bartel via mailop <tom.bar...@gmail.com>
is rumored to have said:
Peter,
Thanks for this thoughtful response - it is good feedback and we'll
take it
to the team here to discuss.
If anyone else has similar input on cut line for a non-commercial
usage,
we'd love the input.
My personal server sees 500-2000 connections (postscreen CONNECT
messages logged, as Peter used) per day, 28014 in the past 29.5 days.
That's a huge drop over the past decade, as some of the bots that used
to make thousands of connections each every night have either become
better behaved or they are isolated to IP-blocked networks. That
includes many connections that never get to DATA, where the Validity
lists would be checked via SA if it were worthwhile. I stopped bothering
to check around the time of the last corporate name change when I did an
analysis and found them entirely valueless for my mail stream.
I did the same analysis for a commercial (SMB domains) mail system I
administer and found essentially the same thing. Validity certified some
senders of pure spam and some senders of pure ham. There was no useful
correlation of any of the Validity checks to whether mail was spam or
ham, although there was a weak correlation of their "safe" senders being
more likely to send spam than the median mixed source. Absolutely not
worth using, especially as it seems the "safe" criteria allow a lot of
spam.
With all that said, the Validity checks and rules remain in SpamAssassin
despite complaints because historically some people have found them
useful and may have created local meta-rules that we won't break
impulsively.
--
Bill Cole
b...@scconsult.com or billc...@apache.org
(AKA @grumpybozo@toad.social and many *@billmail.scconsult.com
addresses)
Not Currently Available For Hire
_______________________________________________
mailop mailing list
mailop@mailop.org
https://list.mailop.org/listinfo/mailop
